+++ This bug was initially created as a clone of Bug #1852056 +++ Description of problem: Currently the etcd health check timeout inside kube-apiserver is hard coded to `2s`. This is very aggressive, change the timeout to 10s This should help the upgrade failures we are seeing in Azure - https://bugzilla.redhat.com/show_bug.cgi?id=1852047
$ git clone https://github.com/openshift/origin.git $ git pull $ git checkout remotes/origin/release-4.5 Previous HEAD position was f093c9b50b Merge pull request #25218 from tkashem/monitor HEAD is now at 6025c28ad9 Merge pull request #25226 from openshift-cherrypick-robot/cherry-pick-25219-to-release-4.5 $ find . -name etcd3.go ./vendor/k8s.io/kubernetes/staging/src/k8s.io/apiserver/pkg/storage/storagebackend/factory/etcd3.go $ grep '10\*' ./vendor/k8s.io/kubernetes/staging/src/k8s.io/apiserver/pkg/storage/storagebackend/factory/etcd3.go ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second) The changes has been merged 4.5 branch.
Checked the PR which only has one change from 2 seconds to 10 seconds, detail see #c1. No test action from QE. Thus could move to VERIFIED.
Confirmed the PR has been shipped to the latest OCP 4.5 payload, see below, $ oc adm release info --commits registry.svc.ci.openshift.org/ocp/release:4.5.0-0.nightly-2020-07-02-002923 | grep origin hyperkube https://github.com/openshift/origin 6025c28ad998e48cfe405578ddc8979809ee81a6 $ git log --date local --pretty="%h %an %cd - %s" 6025c28ad | grep '#25226' 6025c28ad9 OpenShift Merge Robot Wed Jul 1 20:50:23 2020 - Merge pull request #25226 from openshift-cherrypick-robot/cherry-pick-25219-to-release-4.5
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:2409